Prediction of Infertility Type in Women Via Stacked Ensemble Model
Authors: GRILLO Elizabeth Oluwakemi, Oyebode , & ALAO Olujimi
Journal: Research Journal of Mathematics and Computer Science
ISSN Number: 25763989
0
Downloads
10
Views
Abstract
This research shown the importance of implementing ensemble machine learning models for predicting women infertility in Nigerian and other geographical locations in the world. It analysed prevalent clinical risk factors that inclined women to infertility and confirmed the predictive accuracy of machine learning classifiers explored for the research. These included Extreme Gradient Boosting (XGBoost), Extremely Randomized Trees (ExtraTrees) and Convolutional Neural Networks (CNNs). Dataset from two hospitals that had more than two decades of health records of women across Nigeria was used. The data was gathered between 2012 and 2022. It contained information of 5,000 women, age between twenty-five and fifty-five years with twenty -six attributes. The factors were scrutinized to ascertain their predictive significance in identifying women at the risk of infertility via supervised machine learning classifiers.
